Protein Folding with the Parallel Replica Exchange Molecular Dynamics Method - Parallel Computing for Bioinformatics and Computational Biology - page 403

scheme. As shown subsequently, the REMD method is a powerful tool for protein
conformational space sampling.

17.3 PROTEIN FOLDING WITH REMD

17.3.1 Two Example Protein Systems

The α -helix system is a 21-residue peptide with a large propensity of forming α -
helical structure in water at room temperature. It contains three Arg residues in its
Ace-A 5 (AAARA) 3 A-Nme sequence (called Fs peptide, where Fs stands for folded
short and Ace and Nme are capping groups) and has been widely described in the
experimental literature [27-30]. The Fs peptide is then solvated in a cubic box with a
side of 43.7Å containing 2660 TIP3P water molecules [14]. The initial configurations
are selected froma1nssimulation at 700 K of 42 identical solvated systems, starting
from an α -helical configuration and random velocities. The helical content of the
starting configurations ranges from 15 to 73%. The solvated systems are subjected
to 500 steps of steepest descent energy minimization and a 100-ps MD simulation at
constant pressure and temperature (1 atm, 300 K). A modified AMBER94 force field
is used with a cutoff of 8.0 Å for nonbonded interactions. Nonbonded pair lists were
updated every 10 integration steps. The integration step in all simulations is 0.002 ps.
The system is coupled to an external heat bath with relaxation time of 0.1 ps. All
bonds involving hydrogen atoms are constrained using SHAKE with a tolerance of
0.0005Å. The Fs peptide is simulated in the temperature range of 275-551 K with up
to 42 replicas. The temperatures of the replicas are chosen to maintain an exchange
rate among replicas of 8-20%. Exchanges are attempted every 125 integration steps
(0.25 ps).
The β -hairpin system, in contrast, is taken from the C-terminus (residues 41-
56) of protein G (PDB entry 2gb1). It has 16 residues with a residue sequence of
Ace-GEWTYDDATKTFTVTE-Nme. The folding time for this hairpin is about 6

s.
With the explicit solvent, the solvated system has 1361 water molecules [single point
charge (SPC) water, with density 1.0 g / cm 3 ] and also three counter ions (3Na + ) for
neutralizing the molecular system, which results in a total of 4342 atoms in each
replica. A total of 64 replicas are simulated for the explicit solvent systems with
temperatures spanning from 270 to 695 K. For the implicit solvent simulations, only
the protein atoms are present and 18 replicas are used with the same temperature
range (less replicas needed for smaller system size). The OPLSAA force field [31],
as well as AMBER94, AMBER96, and AMBER99 force fields, is used. In explicit
solvent simulations, the periodic boundary conditions are used, and the long-range
electrostatic interactions are calculated by the P3ME method, with a mesh size of
36

µ

36 (grid spacing about 1.0Å). A time step of 4.0 fs (outer timestep) is used
for all temperatures using the RESPA [32-34] algorithm. A standard equilibration
protocol is used for each replica. It starts with a conjugate gradient minimization
for each system. Then a two-stage equilibration, each consisting of a 100 ps MD, is
followed. The final configurations of the earlier equilibration are then used as the

× 36 ×